Major project Retail, E-commerce & Consumer Goods Case Studio

Query understanding and product ranking

Inspired by publicly reported work at Flipkart

Understand what a shopper means (category, brand, size, price), then rank products so the right ones come first.

The brief

Flipkart has written about the stages of a search query: cleaning it, understanding intent (which store or category, which attributes) and ranking results, and about suggesting queries as people type. Build that pipeline on open shopping-query data: classify the category, tag attributes, rank products by relevance, and add autosuggest. Good looks like: category accuracy, attribute-tagging F1, NDCG for ranking against BM25, a working autosuggest, and a search page that uses all of it.

Read the case first

What Flipkart has said in public about this kind of work, on its own pages:

Make it your own

Build search for a catalogue you care about: a college library, a farm-supplies store or a second-hand book market. Write your own relevance labels for 200 queries and compare them with what the model learned from the shopping data.

AITeky has no link with Flipkart, which has not reviewed or endorsed this brief. The brief only draws on what the company has published, uses open data rather than its data, and is a learning exercise: don't present your project as Flipkart's work or system.

What you will build

Classify queries into categories
Tag brands, sizes, colours and prices in queries
Rank products with BM25, then a learned re-ranker
Evaluate with NDCG on the ESCI relevance labels
Build autosuggest from query logs
Build a search page and analyse failures

Free, open data to build it with

It takes a minute with your email. AiTeky Pass (₹999/month or ₹9,999/year) is free for your first 30 days.